Biogeochemistry of benzanthracene in an enclosed marine ecosystem

Eric W. Chan , John D. Kessler , Alan M. Shiller , DongJoo Joung , and Frank Colombo. Environmental Science & Technology 2016 50 (6), 3039-3046...
0 downloads 0 Views 1MB Size
(37) Herbes, S. E., Water Res., 11,493-6 (1977). (38) Pierce, R. H., Jr., Olney, C. D., Felbeck, G. T., Jr., Geochim. Cosmochim. Acta, 38,1061-73 (1974). (39) Kopchynski, K., Hunt, C., Graduate School of Oceanography, University of Rhode Island, Kingston, R.I., 1980, unpublished data. (40) Bassin, N. J., Ichiye, T., J . Sediment. Petrol., 47, 671-7 (1977). (41) Malinky, G., Shaw, D. G., “Proceedings of 1979 Oil Spill Conference (Prevention, Behavior, Control, Cleanup)”, American Petroleum Institute, Environmental Protection Agency, and United States Coast Guard, Los Angeles, Calif., 1979, pp 621-3. (42) Hunt, C., Vargo, G. A., Graduate School of Oceanography, University of Rhode Island, Kingston, R.I., 1980, unpublished data. (43) Myers, A. C., J . Mar. Res., 35,609-32 (1977). (44) Conover, R. J., J. Fish. Res. Board Can., 28,1327-30 (1971). (45) Gearing, P. J., Gearing, J. N., Graduate School of Oceanography, University of Rhode Island, Kingston, R. I., 1980, unpublished data. (46) Vargo, G. A., Graduate School of Oceanography, University of Rhode Island, Kingston, R. I., 1980, unpublished data. (47) Van Vleet, E. S., Quinn, J. G., Enuiron. Sci. Technol., 11, 1086-92 (1977). (48) Farrington, J. W., Quinn, J. G., J . Water Pollut. Control Fed., 45,704-12 (1973). (49) Blumer, M., Sass, J., Science, 176,1120-2 (1972). (50) Keizer, P. D., Ahern, T. P., Dale, J., Vandermeulen, J. H., J. Fish. Res. Board Can., 35,528-35 (1978). (51) Mayo, D. W., Page, D. S., Cooley, J., Sorenson, E., Bradley, F., Gilfillan, E. S.,Hanson, S.A., J. Fish. Res. Board Can., 35,552-67 (1978). (52) Teal, J. M., Burns, K., Farrington, J. W., J . Fish. Res. Board Can., 35,510-20 (1978).

search”, Environmental Protection Agency, Washington, D.C., EPA-600/9-79-035,1979, pp 359-81. (18) Elmgren, R., Oikos S u p p l . , 15,112-20 (1973). (19) Gearing, J. N., Gearing, P. J., Lytle, T. F., Lytle, J. S., Anal. Chem., 50,1833-6 (1978). (20) Hoffmann, E. J., Quinn, J. G., Graduate School of Oceanography, liniversity of Rhode Island, Kingston, R.I., 1980, unpublished data. (21) Farrington, J. W., Quinn, J. G., Estuarine Coastal Mar. Sci., 1, 71 -9 (1973). (22) Santschi, P. H., Li, Y. H., Carson, S., Estuarine Coastal Mar. Sci., in press. (23) Mechalas, B. J., Meyers, T. J., Kolpack, R. L., in “The Microbial Degradation of Oil Pollutants”, Ahern, D. G., Meyers, S.P., Eds., Louisiana State University Publication No. LSU-GG-73-01, 1973, pp 67-79. (24) Ehrhardt, M., Blumer, M., Enuiron. Pollut., 3,179-94 (1972). (25) Anderson, J. W., Neff, J. M., Cox, B. A., Tatem, H. D., Hightower, G. M., Mar. Biol., 27,75-88 (1974). (26) Boylan, D. B., Tripp, B. W., Nature (London), 230, 44-7 (1971). (27) McAuliffe, C., J . Phys. Chem., 70,1267-75 (1966). (28) Peake, E., Hodgson, G. W., J . Am. Oil Chem. Soc., 43,215-22 ( I 966). (29) Eganhouse, R. P., Calder, J. A., Geochim. Cosmochim. Acta, 40, 555-61 (1976). (30) Meyers, P. A., Quinn, J. G., Nature (London), 244, 23-4 (1973). (31) Meyers, P. A., Oas, T. G., Enuiron. Sci. Technol., 12, 934-7 (1978). (32) Zurcher, F., Thuer, M., Enuiron. Sci. Technol., 12, 838-43 (1978). (33) Schultz, D. M., Quinn, J. G., Org. Geochem., 1,27-36 (1977). (34) Boehni, P. D., in “The Tsesis Oil Spill: A Cooperative International Investigation”, Asko Laboratory, Sweden, 1979. (35) Boucher, F. R., Lee, G. F., Enuiron. Sci. Technol., 6, 538-43 (1972). (36) Lotse, E. G., Graitz, D. A., Chesters, G., Lee, G. B., Newland, L. W., Enuiron. Sci. Technol., 2,353-7 (1968).

Received for reuiew November 13,1979. Accepted April 7,1980. Research supported by Environmental Protection Agency Grant No. R803902020.

Biogeochemistryof Benzanthracene in an Enclosed Marine Ecosystem Kenneth R. Hinga’“ and Michael E. Q. Pilson Marine Ecosystems Research Laboratory, Graduate School of Oceanography, University of Rhode Island, Kingston, R.I. 0288 1

Richard F. Lee Skidaway Institute of Oceanography, Savannah, Ga. 31406

John W. Farrington, Kjell Tjessem,2 and Alan C. Davis Department of Chemistry, Woods Hole Oceanographic Institution, Woods Hole, Mass. 02543

Carbon-14 labeled benz[a]anthracene was introduced into a large-scale marine microcosm containing an ecosystem functioning in a manner that was in many respects typical of the shallow coastal waters of the northeastern United States. The radiotracer label made it possible to follow the removal of benz[a]anthracene and labeled metabolites from the water column, their appearance and mixing into the sediments, and the production of 14C02 for 230 days. A t the end of the experiment, 29% had been respired to COz, while the remaining extractable activity (43%)was evenly divided between parent compound and intermediate metabolic products. The biogeochemistry of polycyclic aromatic hydrocarbons (PAHs) in coastal and estuarine ecosystems is important because of the deleterious effects some of these compounds may have on natural resources and human health (1-3).PAHs are released to the environment by a variety of processes such as 1 Present address, Lamont-Doherty Geological Observatory, Columbia University, Palisades, N.Y. 10964. Present address,,University of Bergen, Department of Chemistry, N-5014 Bergen, Norway.

1136

Environmental Science 8 Technology

oil spills, combustion of fossil fuels, forest and grass fires, and industrial effluents (4-8). Analyses of cores containing a historical record of near-shore marine and lake sediments (9-13) indicate increases in concentrations of PAHs by as much as an order of magnitude from ca. 1850 to the present. We can expect that fossil fuel combustion as currently practiced will continue to cause increasing levels of PAHs in the marine environment, especially if there is increased coal combustion in the next century ( 4 , 5 ) .Thus, it is important to understand the biogeochemistry of polycyclic aromatic hydrocarbons in coastal and estuarine areas in order to provide a basis for an evaluation of their long-term effects on these ecosystems. The approach utilized in this experiment was to introduce a 14C-labeled PAH, benz[a]anthracene, into a large marine microcosm. The concentrations of labeled parent compound, some of its metabolites, and 14COz generated from the breakdown of the labeled compounds were measured in various compartments of the ecosystem in the microcosm tank. Since these microcosms contain functioning marine ecosystems (14-1 7), the study was conducted under reasonably natural conditions, with natural populations, for a period of

0013-936X/80/0914-1136$01.00/0

@ 1980 American Chemical Society

time approaching 1 year. It was possible to account for essentially all the labeled compound added to the system. Benz[a]anthracene was chosen because it occurs in many crude oils and refined products, is a major constituent of PAH in combustion products, and is on the Environmental Protection Agency's list of unambiguous pollutants ( 6 , 8 ) . Experimental A large microcosm, one of 12 maintained by the Marine Ecosystems Research Laboratory (MERL) of the University of Rhode Island, was used for this study. Each microcosm is a fiberglass tank 1.8 m in diameter and 5.5 m in height, containing 13 m3 of water and a 30-cm layer of sediment (15).The sediment in the tanks came from central Narragansett Bay and included a soft bottom community dominated by the polychaete Mediomastis ambesita and the bivalve Nucula anulata. During normal operation, approximately 120 L of water from adjacent Narragansett Bay (2€&32%O salinity) was pumped by a diaphragm pump into each tank four times daily, resulting in a washout rate of 3.7% per day. (This is a procedural improvement over the continuous flow regime noted in ref 15. Flow rates were checked weekly.) The tanks were mixed for 2 h, four tirnes a day, to simulate tidal currents and the turbulence regime in Narragansett Bay. The MERL inicrocosms have, over several annual cycles, simulated adjacent Narragansett Bay with respect to biotic structure and chemical concentrations, processes, and rates (14-17). The enclosed ecosystems are large enough to allow many samples to be taken without noticeable depletion, and containment of the ecosystem makes it possible to identify the fate of nearly all the added material. On May 29, 1978, approximately 200 pCi of 12-[14C]benz[alanthracene c'Amersham Corporation, 49 mCi/mmol) was introduced into a MERL tank within an oil-water mixture. This mixture was prepared by shaking 20 mL of no. 2 fuel oil with 17 L of bay water. After 30 min of settling, the bottom 15 L was transferred to another bottle, and -0.93 mg of [14C]benz[a]anthracene in 19 mL of acetone was added to the oilwater mixture. The bottle was shaken and the mixture introduced into the tank at a depth of 2 m during a mixing cycle. Absorption of benz[a]anthracene onto the surfaces of the containers used during the spike resulted in some loss prior to addition into the tank; therefore, the estimate of the amount introduced was obtained by extrapolating concentrations in the tank water back to the time of addition. An identical oil addition except for the [14C]benz[a]anthracene-acetone mixture was made in another MERL tank. Gas chromatographic analysis (18)of the water in this tank 1 h after the addition of the oil-water mixture indicated that -1 g of oil had been added. Sample Colllection and Direct Counts. All samples except cores were taken during a mixing cycle, to ensure that the water column was homogeneous during sampling. Four liters of water was siphoned into glass bottles and 20 mL of hexane was added; the mixture was stirred for 5 h and then allowed to sit overnight. After the hexane layer was drawn off with a pipet, 50 mL of chloroform was added, the mixture was stirred for 4 h and allowed to settle overnight, and the chloroform layer was removed. Aliquots (1-2 mL) of the hexane extract samples were placed in 10 mL of scintillation cocktail (Aquasol, New England Nuclear) and counted for 14Cactivity in a Beckman Model LS-3150T liquid scintillation counter a t MERL. Duplicate samples were counted at Woods Hole Oceanographic Institution (WHOI) by adding appropriate aliquots of the extract to Aquasol scintillation cocktail and counting in a E!eckman Model LS-1OOC liquid scintillation counter. Chloroform samples were also counted at WHOI by the same proceclure. Counting efficiencies in all cases were 90% or better. MERL samples were prepared immediately after

collection. WHOI hexane samples were stored at -15 "C, while chloroform samples were kept refrigerated. Samples of particulate material were obtained by filtering 1or 2 L of water through a 47-mm diameter Gelman A/E glass fiber filter. The filter was placed into 10 mL of cocktail and counted. Phytoplankton samples were obtained by a reverse flow concentration (19, 20) of 2 L of water down to 3 mL through a 10-pm nllon screen. One milliliter of the reverse flow concentrate was placed directly in Aquasol and counted. Zooplankton samples from 350 L of water were obtained by taking a single verticil tow the height of the tank with a 30-cm diameter no. 20 net 180 pm mesh). The volume of the sample was measured and an aliquot (approximately 30%) was filtered through a glass fiber filter and counted. Samples were prepared immediately for counting or stored frozen until processed at WHOI. Half-liter samples of water were transferred to a bottle fitted with two suspended wells and acidified to a pH of ‘262 U o

55%

:p

4

I26

57% 63%

i p 4

126

220

;

220F 4 D

220

;

220F

Figure 6. Distributionof I4Cactivity with depth in sediment cores taken at intervals through the experiment from day 10 to 220. The numbers given in some cases to the right of the profiles at I-cm intervals show the percentage of the I4C activity in each interval which ran with benz[a]anthracene during thin-layer chromatography

Table It. Percentage of the Total I4C Activity Added to a MERL Tank as [14C]Benz[a]anthracene and Recovered as Parent Compound, I4CO2,and Intermediate Metabolic Products day

1 86 126 216

benz[a]anthracene

CO2

1 26 28 29

34 29 21

lntermedlate products a

total accounted for

23 23 22

94 91 83

a Includes, after day 86, data from sediments only, as water column values were then negligible. '' All CO2 was in the water column. Values include estimate of CO2 washed out and COP lost to atmosphere by air-water exchange. Includes the amount of [ 14C]benz[ alanthracene and intermediate metabolic products, estimated ai 11% of total initial activity, washed out of the tank during approximately the first 20 days of the experiment,

+

trapolating the hexane chloroform extract line from Figure 1 to the time of the spike. The [14C]wateris the summation of 14C activity in the hexane chloroform extracts plus the Z14C02.The washout for each day is the 14C activity in the water times the portion of the water removed by washout. The Z[14C]washout is the summation of the calculated daily washouts. The I4CO2 loss to the atmosphere (and to biological fixation) was calculated daily in the same manner as was done to estimate the l'IC02 production. Benz[a]anthracene lost to the atmosphere is thought to be negligible (28). The results of these calculations (Table 111) indicate that the mean total estimated recovery was 80% of the amouht added, and there was no evident trend in this value with time. The 5 to 40% of the 14C activity that was apparently missed during t h P various samplings was probably in the form of benz[a]anthracene or its metabolites not extracted from the sediments or, in the early days of the experiment, from the water column. No absolute calibration of the extraction efficiencies from either medium was attained in these experiments. However, the mean extraction efficiency from water column particulate material was 63% (Table I). Some activity may also have been adsorbed on the tank walls, although in experiments with other hydrocarbons this has not been a significant factor (18). At present we cannot evaluate these possible sources of error.

+

Discussion Both the microbial and benthic animal populations of the microcosm originated from a coastal site near populated areas; therefore, they must have had prior exposure to benz[a]anthracene (8, 10). Quantitative gas chromatography-mass spectrometry measurements of benz[a]anthracene in control MERL sediments gave concentrations of 200-300 ng/g dry weight. Most of the benz[a]anthracene and other PAHs in the sediment likely originate from a combination of low-level chronic inputs of petroleum hydrocarbons and fossil fuel combustion products (8, 10). Thus, these systems have probably received anthropogenically enhanced inputs of benz[a]anthraceme for more than 100 years, in addition to lower level inpu1;s from natural fires since the estuary was formed (7,8,10).If all the approximately 0.93 mg of benz[a]anthracene added during the experiment described above had been incorporated into the sediment, the increase in concentration would have been 27 ng/cm3 in the top 1cm of sediment, about one-sixth of the existing concentration. Thus, the added benz[a]anthraceae did not greatly change the prior concentration. The total 14C activity, including both the parent benz[a]anthracene and Labeled breakdown products, was removed

Table 111. Mass Balance of I4C Activity Added as [14C]Benz[a]anthraceneto One MERL Tank a

day

found In water column

calcd cumulatlve loss through washout and gas exchange

4 7 10 60 86 126 216 220

29 20 15 4 4 2 1 1

8 10 12 31 33 37 40 40

found In sedlment (range)

total accounted for

38 32 (26-37) 62 (37-86) 40 (3 1-48) 58 (50-65) 52 (45-65) 39 (27-56) 28 (19-44)

mean

75 62 89 75 95 91 80 69 80

All values expressed as a percent of the initial estimated value of 31 000 dpm/L. a

from the water with a half-life of about 52 h, while the [14C]benz[a]anthracene itself had a half-life of 24 h, a behavior similar to that of benz[a]anthracene in another microcosm experiment (29).Based on knowledge of the general microbial and animal metabolism of aromatic hydrocarbons and photochemical reaction of benz[a]anthracene,it seems likely that a combination of microbial and photochemical processes gave rise to the labeled products which were found in the water and in the sediments. After a few days, animal metabolism and excretion might also have produced some metabolites (30-34). The presence of phenol and quinone metabolites in the sediment and in the water during the early stages of the experiment demonstrates that the metabolism of benz[a]anthracene to C02 was not rate limited by the initial conversion to quinones or phenols. Once they were formed they survived long enough to be detected in substantial quantities, up to 23% of the initial total 14C activity. Benz[a]anthracene, having a low water solubility and low rate of metabolism in the water column, became associated with sediments either through direct adsorption on the bottom after turbulent mixing of water to the sediment-water interface or by adsorption onto or incorporation into suspended particles and subsequent deposition. Once incorporated into surface sediments, the benz[a]anthracene and its metabolites were mixed deeper into the sediments by benthic animal activity. Both the removal a n d mixing processes took place on a time scale which was faster than the conversion rate of benz[a]anthracene and extractable metabolites to nonextractable metabolites and C02, as the total 14Cactivity in the sediment did not decrease rapidly. Because of the rapid transfer of the benz[a]anthracene to the sediment, it is not possible to specify whether the initial high rate of metabolite production was carried out in the water column on suspended particulates or in the sediment. However, studies of [14C]benz[a]anthracenein bottle experiments (35)using water and sediments from both Narragansett Bay and the MERL microcosms indicated that no 14C02production occurred in water, while a measurable 14C02production took place in sediments. The greater percentage of 14C-labeled metabolites in the top centimeter of sediment than in the 1-2-cm layer indicates that the metabolism of benz[a]anthracene occurred more rapidly in the top layer. For other coastal sediments it has been shown that microbial degradation of petroleum hydrocarbons is more rapid near the surface than in lower layers (36). In the MERL sediments this may be due in part to the reducing conditions below 1cm, while the surface sediments are oxidized (37).The increased 14Cactivity found in polar metabolites, relative to that in the parent compound, deeper in the core (4-5 cm layer) provides some Volume 14, Number 9, September 1980

1141

insight into diagenetic processes and suggests further studies. This observation could perhaps be explained by assuming a greater diffusive mobility of the polar compounds, compared to benz[a]anthracene, or by the action of benthic animals ingesting and partially metabolizing the benz[a]anthracene and then excreting the metabolites at depth. The variability of the depth distributions for quadruplicate cores on days 216 and 220 illustrates that, despite the passage of 200 days since the introduction of the benz[a]anthracene to the system, the combination of metabolic processes and biological and physical mixing of sediments had not resulted in a uniform distribution of either parent compound or extractable metabolites in the upper 4 to 5 cm of sediment. This is consistent with the view that in benthic ecosystems of this type there is a dynamic interaction of fluid and particle movements brought about by different benthic animals with varying feeding strategies and burrowing activities, resulting in local patchiness of chemical distributions. The data do not allow an unequivocal calculation of how long it would take for the remaining 14C activity to be converted to 14C02.The 14C02production followed neither a linear nor an exponential loss rate (Figure 5). We do not know the mechanisms that control the rate of benz[a]anthracene breakdown, so we have no basis to choose a method of calculation. If the 14C02production rate were to remain the same as during the last 100 days of the experiment (Figure 5), all the labeled benz[a]anthracene would be converted to W02 within 3.5 years after the spike. Since the temperature of the sediments would be greater in the summer, it may be more reasonable to use the decrease in 14Cactivity in sediment cores from day 10 to day 216, which includes warmer temperatures, to predict 2.0 years for the complete mineralization of the benz[a]anthracene. Assuming an exponential loss rate and depending on the interval chosen, half-lives of 1.2 to 3 years may be calculated. The occurrence, however, of benz[a]anthracene at some depth in natural sediments suggests that a fraction of this compound and perhaps some of its metabolites may persist indefinitely. Because so many of the better known chemical functions in the MERL ecosystems follow those in the adjacent bay, both in magnitude and timing, it seems reasonable to suggest that the behavior of benz[a]anthracene in the microcosm would be similar to that in Narragansett Bay and similar coastal bodies of water. Nevertheless, with the present state of our knowledge, it would be risky to extrapolate from these data to the situation in other environments. Temperature, water-sediment exchange processes, rates of bioturbation, physical characteristics of the sediment, and other effects may help control the benz[a]anthracene degradation rate. If the addition had been in the winter and then progressed to the warmer months of summer, the benz[a]anthracene loss through microbial and animal metabolism might have been different. During the winter when there is less rapid bioturbation, the benz[a]anthracene may remain for a longer period in the surface sediments. In regions of rapid sedimentation, or in areas subjected to destruction of the biota by an oil spill, benz[a]anthracene buried in the sediment may remain in anoxic layers and be protected to some extent from microbial attack. The persistence of metabolites and reaction products raises important questions with regard to assessing the environmental risk associated with levels of aromatic hydrocarbons in coastal ecosystems. Many of the metabolites and reaction products of aromatic hydrocarbons are thought to be mutagenic and/or carcinogenic (31).Measurements of the loss of aromatic hydrocarbons after a spill do not provide a measure of the metabolites which may persist. This is especially true for the heavier molecular weight phenanthrenes and naphthalenes and compounds such as benz[a]anthracene and benzo[a]pyrene, which persist longer than the lower molecular 1142

Environmental Science & Technology

weight benzenes and naphthalenes (22). The persistence of these compounds suggests the presence of a hazard. However, further measurements and experiments are needed to determine to what extent these compounds are available for uptake by biota and subsequent long-term effects.

Acknowledgments We thank Dr. Patrick Gearing, Dr. Juanita Gearing, Dr. James Quinn, Dr. Candace Oviatt, and Mr. David Johnson for helpful discussion and cooperation. Drs. J. and P. Gearing also provided the analysis of total oil, and made several editorial suggestions. We also thank the entire MERL staff for their helpfulness. Literature Cited (1) National Academy of Sciences, “Petroleum in the Marine Environment”, Washington, D.C., 1975. (2) “Proceedings of a Workshop on Scientific Problems Relating to Ocean Pollution”, Estes Park, Colo., July 10-14, 1978, U.S. Department of Commerce, National Oceanic & Atmospheric Administration, Environmental Research Laboratories, Boulder, Colo., 1979. (3) Anderson, J. W., Malins, D. C., J. Fish.Res. Board Can., 35,679 (1978). (4) Farrington, J. W., in “Proceedings of a Special Symposium of the American Chemical Society Environmental Division”, in press. (5) Wakeham, S.W., Farrington, J. W., in “Proceedings of a Special Symposium of the American Chemical Society Environmental Division”, in press. (6) National Academy of Sciences, “Particulate Polycyclic Organic Matter”, Washington, D.C., 1972. (7) Youngblood, W. W., Blumer, M., Geochim. Cosmochim. Acta, 39, 1303 (1978). (8) LaFlamme, R. E., Hites, R. A., Geochim. Cosmochim. Acta, 42, 289 (1978). (9) Farrington, J. W., Frew, N. M., Gschwend, P. M., Tripp, B. W., Estuarine Coastal Mar. Sci., 5,793 (1977). (10) Wakeham, S.G., Schaftner, C., Giger, W., Geochim. Cosmochim. Acta, 44,403 (1980). (11) Hites, R. A,, LaFlamme, R. E., Farrington, J. W., Science, 198, 829 (1977). (12) Grimmer, G., Boehnke, H., Cancer Lett., 1,75 (1975). (13) Muller, G., Grimmer, G., Bohnke, H., Naturwissenshaften, 64, 427 (1977). (14) Pilson, M. E. Q., Vargo, G. A,, Gearing, P., Gearing, J. N., in “Proceedings of the 2nd National Conference Interagency Energy/Environment R&D Program”, EPA-600/9-77-012, 1977, pp 513-6. (15) Pilson, M. E. Q., Oviatt, C. A., Vargo, G. A., Vargo, S. L., in “Advances in Marine Environmental Research”, Jacoff, F. S., Ed., EPA 600/9-79-035,1979, pp 359-81. (16) Pilson, M. E. Q,, Oviatt, C. A., Nixon, S.W., in “Symposium on Microcosms in Ecological Research”. Savannah River Ecoloav -_ Laboratory, 1978, in press. (17) Elmgren, R., Vargo, G. A., Grassle, J. F., Grassle, J. P., Heinle, D. R., Langlois, G., Vargo, S.L., in “Symposium on Microcosms in Ecological Research”, Savannah River Ecology Laboratory, 1978, in press. (18) Gearing, J. N., Gearing, P. J., Wade, T. L., Quinn, J. G., McCarty, H. B., Farrington, J. W., Lee, R. F., in “Proceedings of the 1979 Oil Sui11 Conference”. American Petroleum Institute. Washington. D.C., 1979, pp 555-64. (19) Dodson, A. N., Thomas, W. H., Limnol. Oceanogr., 9, 455 (1964). (20) Hinga, K. R., Davis, P. G., Sieburth, J. McN., Limnol. Oceanogr., 24,536 (1979). (21) Bligh, E. G., Dyer, W. J., Can. J . Biochem. Physiol., 37, 911 (1959). (22) , . Teal. J. M.. Burns. K. A.. Farrington. . J. W., J . Fish.Res. Board Can., 35,510 (1978).’ (23) Hinga, K. R., Marine Ecosystems Research Laboratory, Graduate School of Oceanography, University of Rhode Island, unpublished data. (24) Adler, D., Amdurer, M., Santschi, P. H., in “Symposium on Microcosms in Ecological Research”, Savannah River Ecology Laboratory, 1978, in press. (25) Santschi, P. H., Li, Y. H., Carson, S., Estuarine Coastal Mar. Sci., in press. (26) Rhoads, D. C., Mar. Biol. Annu. Rev., 12, 263 (1974). (27) Rhoads, D. C., McCall, P. L., Yingst, J. Y., Am. Sci., 66, 577 (1978). L

.

(28) Southworth, G. R., Bull. Enoiron. Contam. Tonicol., 21, 507 (1979). (29) Lee, R. F., Gardner, W. S., Anderson, J. W., Blaylock, J. W., Barwell-Clarke,J.,Environ. Sei. Technol., 12,832 (1978). (30) Stegeman, J. J., Binder, R. L., Biochem. Pharmacol., 28,1686

(1979). (31) Stegeman, J. J., J . Fish. Res. Board Can., 35,688 (1978). (32) Malins, D. C., in “Fate and Effects of Petroleum Hydrocarbons in Marine Organisms and Ecosystems”, Wolfe, D. A., Ed., Pergamon Press, New York, 1977, Chapter 5, pp 47-59. (33) Corner, E. D. S., Ado. Mar. Biol., 15,289 (1978). (34) Lee. R. F., in “Proceedingsof the 1975 Conference on Prevention and Control of Oil Pollution”, American Petroleum Institute, Washington, D.C., 1975, pp 549-53.

(35) Lee, R. F., Skidaway Institute of Oceanography,Savannah, Ga., 1978, unpublished data. (36) Gardner, W. S.,Lee,R. F.,Tenore, K. R., Smith, L. W., Water, Air Soil Pollut., 11,339 (1979).

(37) Pilson, M. E. Q., Marine Ecosystems Research Laboratory, Graduate School of Oceanography, University of Rhode Island, 1978, unpublished data.

Received for review January 11,1980. Accepted April 14,1980. This work was supported by Environmental Protection Agency Grant R 80607202. This paper is Woods Hole Oceanographic Institution Contribution No. 4513.

NOTES

Correlation of Microbial Degradation Rates with Chemical Structure N. Lee Wolfe’, Doris F. Paris, William C. Steen, and George L. Baughman Environmental Research Laboratory, U.S. Environmental Protection Agency, Athens, Ga. 30605

1 Structure-reactivity relationships are established for the

microbial degradation of selected organic compounds. Second-order microbial degradation rate constants determined in natural water [samples for six compounds correlate with the second-order alkaline hydrolysis rate constants. Also, second-order microbial degradation rate constants for four phthalate esters obtained with organisms from sedimentwater samples c’orrelate with the second-order alkaline hydrolysis rate constants. Similar correlations are demonstrated for the rates of oxidation of substituted phenols by mixed microbial cultures isolated from soils and Hammett u constants. Correlations Eietween rate constants for specific chemical reactions and values for a selected physical or chemical property of the organic compounds are well established in the chemical literature (I ). Such correlations can be used to predict rate constants required to assess the behavior of organics in aquatic environments (2, 3). Similar correlations of enzyme-catalyzed reactions, although often more complex, have also been demonstrated ( I , 4 ) . Several studies have addressed the effect of chemical structure on susceptibility to microbial breakdown using bacteria isolated from soils (5,6)and natural waters (7, 8 ) , but, unfortunately, the correlations have not been successful. The major difficulty with these studies has been the choice o f a measurement of biological reactivity to use in the correlation. Recent studies of the kinetics of microbial degradation of organics in natural water samples provide a data base of rate constants that can be used to examine the relationship of chemical structure to microbial degradation. In these studies, Paris and co-workers (9-11) applied a second-order rate expression to describe the rate of disappearance of certain organics mediated by microorganisms in natural water samples. Linear regression analysis using the second-order rate constants, k b (L erg-' h-l), determined in these investigations and the corresponding second-order alkaline hydrolysis rate ~ s-l), reported by Wolfe and co-workers constants, h o (1W-l (11-15) gives the correlation described by the linear equation (16):

+

log k b = m log h o ~c

(1)

Values of m and c along with standard errors of the estimates and the correlation coefficient, r 2 , are given in Figure 1. (The second-order biolysis and hydrolysis rate constants for methyl benzoate and anisate according to Wolfe and Paris (11) are and 5.4 X L org-I h-’ and 5.0 X lov3and 1.1 7.0 X X M-l s-l, respectively.) For this group of compounds, 97% of the variance in the data is accounted for by alkaline hydrolysis. In addition, the F value is 141, which is significant at the greater than 99% confidence level (CL). The correlation (R’) is improved by inclusion of a second parameter, the octanol--water partition coefficient (KO,) (17 ) ,which has been postulated to be proportional to the binding strength of a compound at the reaction site ( 4 ) or proportional to biosorption by organisms (18).(Octanol-water partition coefficients were calculated using the transposed linear equation, log K O , = -0.653 logs 0.880, obtained by linear regression analysis of the data in ref 17.) The resulting equation is: log k b = m log OH n log KO, c (2) With the inclusion of KO, as a dependent variable, the multiple linear regression coefficient, R 2 ,is 0.988, which accounts for an additional 2% of the variance of the data. The values of m , n, and c are 0.53 f 0.03,0.13 f 0.06, and -11.8 f 0.2, respectively. However, the F test shows that inclusion of K O , as a dependent variable after KOHis not significant. A similar correlation is also shown for phthalate esters. The second-order disappearance rate constants reported by Steen et al. (19), for the microbial degradation of four phthalate esters by organisms from sediment samples, and the secondorder alkaline hydrolysis rate constant for the corresponding esters reported by Wolfe et al. (20) were correlated by use of Equation 1. (Reference 19 contains the biolysis rate constant for di-n-butyl phthalate and a description of the experimental methodology and calculations. Second-order biolysis rate constants for the dimethyl, di-n-octyl,and bis(2-ethylhexyl) 3.1 X and 4.2 X L erg-' phthalates are 5.2 X h-’, respectively.) Linear regression analysis of the data gives the values shown in Figure 2 ( r 2 = 0.933). In addition, the F value is 27.7, which shows that the correlation is significant at the 95% CL. Although 93% of the variance is accounted for by alkaline hydrolysis, the correlation is improved by using Equation 2 ( R 2= 0.994). Thus, the octanol-water partition coefficient as an independent variable accounts for an addi-

This article not subject to U.S. Copyright. Published 1980 American Chemical Society

+

+

+

Volume 14, Number 9, September 1980

1143